Open oybed opened 6 years ago
@oybed So the initial culprit seems to be here: https://github.com/openshift/openshift-ansible/blob/master/roles/openshift_openstack/tasks/provision.yml#L1-L16
provision.yml
skips running generate-templates.yml
whenever we set state=absent. However that is where stack_template_path
is set, which causes everything to fail out whenever when it runs the third step (Validating the heat template).
So the question now is whether we craft something to work around this for the time being. Or if we try to get this fixed in openshift-ansible. Wanted to get your thoughts before digging in any further.
@tylerauerbeck yeah, it's a bit of a challenge to utilize the "downstream" code for this, so we may just need to move it back into casl
, but still keep it in sync (i.e.: submit updates to openshift-ansible
) as we go. Let's discuss before we do anything further on this.
What's the status of this? @tylerauerbeck @oybed
Im hitting the same issue in 3.11 with the latest casl-ansible container image:
TASK [../../../galaxy/openshift-ansible/roles/openshift_openstack : validate the Heat template] *******************************************************************************************************************
task path: /tmp/src/casl-ansible/galaxy/openshift-ansible/roles/openshift_openstack/tasks/provision.yml:42
Tuesday 18 December 2018 11:35:08 +0000 (0:00:01.567) 0:00:43.679 ******
fatal: [localhost]: FAILED! => {
"msg": "The task includes an option with an undefined variable. The error was: 'stack_template_path' is undefined\n\nThe error appears to have been in '/tmp/src/casl-ansible/galaxy/openshift-ansible/roles/openshift_openstack/tasks/provision.yml': line 42, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: validate the Heat template\n ^ here\n"
}
@jtudelag this is a low priority issue as you can just use the openstack stack delete
command (or the OSP web console).
@jtudelag one additional piece of info - before you delete the cluster with the above mentioned command, make sure any PVs are deleted to avoid leaving Cinder volumes behind (+ double check the OSP tenant afterwards that all related volumes were deleted).
@oybed ok. Btw, there a few non-attached volumes already.
Also, deleting the stack manually would leave some trash, such as DNS records and Host registered in Satellite.
@jtudelag the the DNS, host registration., etc. isn't handled by the delete playbook anyway. That's a different issue. The openshift-ansible
delete functionality takes care of (some) of that.
@oybed So how do you manage that today? Manually? I mean, removing DNS records & Sat Hosts.
After moving to use
openshift-ansible
as our source for the OCP on OSP deployments, the delete a cluster option doesn't work. A PR is needed to update the variables passed to theopenshift_openstack
role to make it happy.